This project examines the contribution of lexically encoded participant information to the construction of a representation of the meaning of a sentence or discourse. Its overarching goal is to illuminate foundational aspects of how lexically encoded participant information is mentally represented and how it influences sentence and discourse comprehension mechanisms. The proposed research will use on-line experimental methods that are sensitive to immediate comprehension processes to address this goal. Some of the proposed research is aimed at developing and applying a behavioral measure to identify participant information that is encoded in the lexical representation of different classes of verbs. The resultant detailed picture of the verbal mental lexicon will be used to predict and test how sentences with different types of participant information are processed. Other portions of the project are aimed at uncovering the discourse properties of unexpressed verb argument information (i.e. implicit arguments) and their influence on the establishment of discourse coherence. This program of research should yield a significant body of knowledge with extensive ramifications and relevance to further investigations of adult normal and neurogenically disordered sentence comprehension, and to the study of first and second language acquisition.